Table 18. — Estimates of the number of yearling male seals, based on tag recoveries, 

 year classes 1961, 1962, 1964, and 1965, Pribilof Islands, Alaska 



1 Tags from animals of unknown age were divided on the basis of the proportion 

 observed among animals of known ages. 



2 N=(MC)/R where M is number tagged, C is the kill, and R is the number of tags 

 recovered. 



3 We assume an error of 0.032 in classifying seals by age during tagging. 

 * We assume an error of 0.059 in classifying seals by age during tagging. 



the right upper canine teeth of those taken 

 in the kill. For the 1961 and 1962 year 

 classes the error in classification by age 

 at the time of tagging was 3.2 and 5.9 



percent, respectively. The data for the 1964 

 and 1965 year classes are incomplete; we 

 assumed an error of 5.9 percent for these 

 year classes. 



FORECAST OF THE KILL OF MALE SEALS IN 1968 



A comprehensive review of the prediction 

 methods that have been used over the past 

 several years has been reported (U.S. Fish 

 and Wildlife Service, Bureau of Commercial 

 Fisheries, 1969). The most reliable method of 

 forecasting the kill of 4-year-olds has been 

 based on the kill of 3-year-olds from the year 

 class, together with some measure of the timing 

 of their return to the islands at age 3. The esti- 

 mate of timing depends upon the kill per round 

 (5-day period), which is affected by the behavior 

 of the animals and by the management sched- 

 ule. Management changes begun in 1967 have 

 cast doubt on the validity of the measure of 

 timing for the 1968 forecast. There is some un- 

 certainity concerning the prediction of the kill 

 of 4-year-olds, especiallybecause the estimate 

 of the number of yearlings from the 1964 year 

 class suggests a higher kill. 



Earlier reports referred to the use of esti- 

 mates of the yearling population as a basis for 

 forecasting, under the theory that most of the 

 compensatory and variable mortality occurs 

 during the first year of life. In other words, 

 the survival rate after age 1 should be rea- 

 sonably constant. Therefore, if good estimates 

 can be obtained of the yearling survivors and 

 of this constant survival rate from age 1 to 

 ages 3 or 4, the problem of forecasting would 

 be minimal. 



Tagging of yearlings provides a marked 

 cohort of known size in the 1-year-old group 

 but this does not yield a population estimate 

 until a random sample of this group has been 

 subsequently obtained. The first such sample 

 becomes available when these seals are killed 

 at age 2, but so far the estimates from tag 

 recoveries of 2-year-old males have been 



18 



